1 |
On Sat, Aug 25, 2018 at 10:11 PM Andrew Savchenko <bircoph@g.o> wrote: |
2 |
> |
3 |
> On Sat, 25 Aug 2018 14:24:02 -0400 Mike Gilbert wrote: |
4 |
> > PMS does not specify that behavior (skipping src_test with USE=-test). |
5 |
> > It is better to define the requrement explicitly rather than relying |
6 |
> > on a Portage-specific behavior. |
7 |
> |
8 |
> Then PMS should be fixed. Putting useless code in thousands |
9 |
> of ebuilds due to bureaucratic reasons is ridiculous. Having strict |
10 |
> conformance to the PMS is good, but common sense should still be |
11 |
> considered. |
12 |
|
13 |
I disagree with your opinion. Adding one line of code to many ebuilds |
14 |
is not a crazy idea, and it doesn't need to happen overnight. Adding |
15 |
RESTRICT="!test? ( test )" is certainly not harmful. |
16 |
|
17 |
The advice given in the devmanual should agree with PMS. I think PMS |
18 |
is unlikely to change; if it does, the devmanual can be updated to |
19 |
match. |
20 |
|
21 |
Personally, I would not like to see a change in PMS in this regard. I |
22 |
think controlling tests via USE conditionals in RESTRICT is a nice way |
23 |
to avoid hard-coding the meaning of individual USE flags into PMS. |
24 |
It also gives the ebuild author more flexibility. For example, he |
25 |
could put something like this in RESTRICT if some exotic "foo" feature |
26 |
causes tests to fail: |
27 |
|
28 |
RESTRICT="foo ( test )" |